We’re hiring! The University of Arkansas Department of Education Reform is seeking a tenure-track Assistant Professor of Education Policy. Join a dynamic research community committed to rigorous education policy research and doctoral training. Apply: shorturl.at/DlpHI Please share widely!

The robots did come. Now AI has arrived too. But the skills that matter have changed less than the technology: reading, math, critical thinking, communication, adaptability, judgment, and working with others. The real question: can education change fast enough to keep up? shorturl.at/liFYm

Fewer teachers, more health professionals Source: Education Pays, by College Board research.collegeboard.org/media/pdf/ed...

📢 New @Glabor_org working paper: Education and Earnings in Europe Using compulsory schooling reforms and comparable microdata across 16 European countries, we estimate the causal returns to education. ideas.repec.org/p/zbw/glodps... #EconTwitter #EducationEconomics #HumanCapital

How College Education Divided America Great piece by @dacemoglumit.bsky.social based on his new book What Happened to Liberal Democracy? (which I just received) time.com/article/2026...

2.25 billion, the number of cups of coffee drunk across the world each day [Source: The Economist]

Education still pays in Arkansas. Our new study finds that each additional year of schooling is associated with 7.5% higher earnings. The return is even higher for women: 8.6%. For university education, the returns rise to 10.8% for women. shorturl.at/99svv

Excited to speak at the 2026 Emerging School Models Conference at @Kennedy_School! I will explore accountability in innovative educational models. Join us on Sept 17-18 to discuss the future of education! Register: bit.ly/4pM32s6 #EducationReform #EmergingSchools #HarvardConference

📢 New publication! Education and Earnings in Arkansas is now published in Journal of Economic Insight We find that each year of schooling is associated with a 7.5% increase in earnings, with even higher returns for women. shorturl.at/99svv #EducationEconomics #Arkansas #HumanCapital

Excited to speak at the 2026 Emerging School Models Conference at @Kennedy_School! I will discuss how choice systems might be at odds with accountability. Join us on Sept 17-18 to discuss the future of education! Register: bit.ly/4pM32s6 #EducationReform #EmergingSchools #HarvardConference

University-for-all harms poor students the most, says @TheEconomist, which also calls for improving quality BEFORE students go to high school. Low standards are probably what keep returns to higher education high, despite the growth in enrollments. shorturl.at/cALW4 shorturl.at/PViiT

How many times a day do you think about Alexander the Great? shorturl.at/99wjS

The paper finds that returns to university education for men rose from 14.3% to 15.0% during the COVID period, largely because less-educated workers suffered much greater employment losses. tandfonline.com/eprint/EJ35U...

Learning Pays More Than Schooling What matters more: how long students stay in school, or how much they actually learn? My latest substack: harryapatrinos.substack.com/p/learning-p...

Education isn't just about years in school—it's about what students learn. A review of 73 studies in 41 countries finds that a 1 SD gain in cognitive skills is associated with ~15% higher earnings. Better learning, better livelihoods. link.springer.com/article/10.1...

The average causal return to an additional year of schooling is about 10%. The biggest gains come from students whose education is expanded by policy: constrained learners who otherwise would not stay in school. www.sciencedirect.com/science/arti...

Go Habs!

Today I presented my recent learing loss COVID-19 paper (www.sciencedirect.com/science/arti...) at the Learning Loss forum (online) organized by the Arab Bureau of Education for the Gulf States, which was opened by the Kuwait Minister of Education. @JakubowskiEvid @TGajderowicz

I will be speaking at the SustainableED 2026 Conference at the Annenberg Institute at Brown University tomorrow, where I will present A Review of Human Development and Environmental Outcomes annenberg.brown.edu/sustainablee...

Human Capital Is Climate Policy open.substack.com/pub/harryapa...

What do we really know about education & climate outcomes? In review with Ambasz & Gupta: 31 studies → 27 observational, only 4 quasi-experimental. Education shifts attitudes, but behavior is harder to change. We need more causal evidence on what actually works. shorturl.at/RmTIM

New paper: Returns to Education in Greece (1977)—post-dictatorship evidence. Using IV (father’s education), I estimate a causal return of 7.9% per additional year of schooling. Strong first-stage + endogeneity tests support the approach. Education pays- and causally. www.kepe.gr/wp-content/u...

Designing Effective PPPs in Education: Does governance design matter more than private provision My presentation from the 2026 AERA meetings hpatrinos.com/2026/04/17/d...

Education yields about 10% returns per year globally. Our new HCER Lab at @ua_edreform is focused on measuring how education investments translate into real economic outcomes. Learn more about the lab: edre.uark.edu/humancapital...

New paper (open access): The learning crisis - 3 years after COVID-19 📉 Achievement −0.11 SD below trend ⏳ Longer closures → bigger losses ⚠️ Low performers hit hardest (−0.23 SD) ⚖️ Gaps widened This is divergence, not recovery. Targeted action needed now. doi.org/10.1016/j.le...

Just out as GLO paper: Returns to Education in the United States A Comparison of OLS and Double Machine Learning Methods w/ Helal & Hiraki www.econstor.eu/bitstream/10...

Since we can trust OLS, then global evidence from 142 economies (853 surveys) shows right-skewed distribution of returns: education more likely than not to increase earnings, with meaningful upper tail. This pattern aligns with how individuals make schooling decisions. emerald.com/ijm/article-...

Investigating the reproducibility of the social and behavioral sciences: > Miske et al examined 600 papers > 72% shared neither data nor code > Reproduction attempted only for 30% of papers Results: 54% precisely reproducible, 3/4 achieved ‘approximate reproducibility’ nature.com/articles/s41...

Four Education Trends that Countries Everywhere Should Know About open.substack.com/pub/harryapa...